by Intel has dropped some fresh details about its new Lunar Lake mobile CPU architecture, including some performance and power claims, but hasn’t backed that up with any details. Claiming that it’s not ready to showcase CPU performance right now, we only have a few promises and an unexpected look at Microsoft Teams as a CPU benchmarking platform.
Lunar Lake represents the next generation of Intel processors for thin and light laptops and is the first modern CPU design to be made entirely outside of its own manufacturing facilities. Instead the Lunar Lake chip will be made entirely by TSMC using the N3B node.
That’s not a bad thing for the end user. After all, we do Seriously care where the silicon is slapped together if it delivers?
And Intel is promising it will, with the new chips also being the first to use the new Battlemage graphics cores. This integrated Xe2 GPU is reported to deliver 50% higher performance than the competing Meteor Lake Core Ultra 7 165U in terms of 3DMark Time Spy numbers.
It is worth noting that 3DMark benchmarks have historically been very kind with Intel’s Arc GPUs, so it’s worth noting that actual gaming performance may still be very unrestricted. And the Core Ultra 7 165U only had four Xe cores inside, too, so it wasn’t exactly the hero chip in terms of Meteor Lake graphics performance.
But it’s also that GPU that helps the Lunar Lake chip deliver AI performance that’s about 40% faster than Qualcomm’s new Snapdragon X Elite chips when it comes to Stable Diffusion.
AI is the battleground for all processor architectures to fight out over the next 12 months at least, so that’s clearly a point Intel wants to hammer home. The Xe2 cores have XMX Battlemage matrix engines built into the iGPU, which alone deliver 60 TOPs. Then you have the Lunar Lake NPU itself, which is promised to bring up 45 TOPS, which means, in total, you get more than 100 TOPS of AI processing brew out of the box.
Clearly the other battleground for any low-power laptop platform is efficiency. Intel is promising that its Low Power Island will deliver twice the compute of Meteor Lake, which should give it real power advantages on simple applications, such as Microsoft Teams. That recognized benchmark standard.
The Low Power Island was first introduced with Meteor Lake, where Intel dropped a few Efficient cores on the SoC away from the actual compute tiles. This allowed the entire tile to be calculated effectively to go to sleep when the device only needed a pair of E-cores to keep the lights on. With Lunar Lake the knowledge is that they will do more than keep the lights on and will be more valuable in terms of calculations.
Intel hasn’t said if that means it’s using more E-cores in the Low Power Island SoC, or if it’s the new Skymont microarchitecture that delivers this extra computing performance with the same number of new E-cores.
Still, using that Team chat benchmark, Intel claims up to 30% lower total package power compared to the Ryzen 7 7840U and up to 20% lower than the Snapdragon 8cx Gen 3. That’s especially no the new Snapdragon X Elite or X Plus chips, as Intel is only making claims against publicly released benchmark data on Qualcomm chips. And it’s strange that Qualcomm hasn’t said how much power they drain in Microsoft’s conference app, although it has said that its lower spec X Plus chips will give twice the battery life against Meteor Lake in terms of Teams.
Still, for his part, Intel’s Bobby Hallock, once of AMD, was his usual scathing self, declaring that “you haven’t seen x86 power characteristics quite like this.”
That’s great for Lunar Lake when it hits laptops in the July – September time frame, as long as it can back up all these claims in real terms when the chips and laptops are out there .
But good things could also be in store for Arrow Lake. That’s the new CPU architecture for next-gen standard laptops and desktops coming later this year and it’s set to use the same Lion Cove and Skymont microarchitectures for its Performance and Efficiency cores respectively as the Lunar Lake. Although they will be reported – at least in part – using Intel’s fancy new 20A production process.
It won’t use the new Battlemage-based Xe2 cores for the relevant iGPUs, though, but you’d expect those chips to be paired more regularly with discrete graphics cards. And we can’t all have nice things, either.
But in real terms, this latest information drop from Intel is a bit light on the details and relies more on promises versus the competition’s claims. How any of the new chips – Intel Lunar Lake, AMD Strix Point, or Qualcomm’s Snapdragon X Elite – will really stack up in the final calculation is going to take time and independent testing to resolve.
Although we probably won’t be using Microsoft Teams as part of our own benchmark suite when it comes to it. I’m sorry, Nick.
